As disability becomes increasingly recognized as a diabetes outcome of major clinical and public health importance, it becomes less clear how diabetes causes disability. The goal of the proposed research is to investigate whether hyperglycemia itself has an independent association with disability in mobility, endurance, strength and activities of daily living. We will test the hypothesis that hyperglycemia contributes to the development of frailty (a syndrome of vulnerability to stressors recently shown to be prevalent in community-dwelling older adults) and through this to disability, via a pathway involving inflammatory mediators. The specific aims are: l) to describe change over time in level of hyperglycemia measured by glycohemoglobin; 2) to evaluate associations of hyperglycemia with inflammatory mediators and how that association is modified by obesity; 3, to evaluate associations of hyperglycemia with characteristics defining frailty (weight loss, weakness, slowness, exhaustion, low exercise tolerance) and the frailty syndrome itself, to determine if hyperglycemia at baseline or its change over time predicts onset or progression of frailty; 4, to evaluate the contribution of hyperglycemia to measures of disability through frailty, and overall, adjusting for comorbidity; 5, to seek evidence for potential critical points in the pathway from hyperglycemia to disability that might be amenable to intervention, such as (a) threshold levels of hyperglycemia; (b) subsets of older people in whom hyperglycemia may be particularly problematic such as those with vascular conditions and obesity; and 6, to evaluate population attributable risk of frailty and disability due to the hyperglycemia frailty pathway. Our research strategy includes: study of women who are both functional and disabled using two complementary studies, the Women's Health and Aging Studies I and II, which have nearly identical measures of relevant variables; and cross sectional and longitudinal data analysis. Analyses will focus on longitudinal change models including random effect growth curves, random-effect change models, and time to event models. This research will link inflammatory disruption in diabetes to that of frailty and disability, help determine the potential impact of hyperglycemia on adverse health outcomes important to older adults, and provide important preliminary data for future studies, observational and interventional, of glycemic control in older adults with diabetes.